The proposed project is a five-year research study designed to refine and test a brief add-on HIV-risk reduction intervention with a multicultural sample of sexually active adolescents in outpatient treatment for alcohol and other drug (AOD) use problems. Adolescent AOD abuse and sexual risk behaviors are significant sources of morbidity and mortality due to the multiple ways that these behaviors increase risk for HIV/STD exposure. Compared with the general population of adolescents, teens undergoing treatment for AOD abuse are more likely to report multiple problem behaviors that include HIV-related and other health-risk behaviors. The proposed study is a randomized clinical trial evaluating the efficacy of Guided Self-Change for reducing HIV-risk behavior (GSC-HIV). Key features of the study include manualized treatments, theory-driven hypotheses, a randomized controlled trial design, and a culturally and ethnically diverse sample of youth with co-occurring problems of AOD abuse and sexual risk behaviors. Adolescent clients (n = 844) attending outpatient treatment programs at a local substance use treatment provider (The Starting Place in Hollywood, Florida) will be randomly assigned to two conditions (1) Treatment as Usual (TAU), or (2) TAU plus a brief motivational, behavioral HIV-risk reduction intervention (GSC-HIV). GSC-HIV will consist of 4 weekly individual sessions of one-on-one counseling with a trained therapist.Participants will be assessed immediately before and after intervention and at 3- 6- and 9-month follow-ups. Participants in the TAU condition will complete a parallel schedule of assessments. Our Primary Hypothesis is that adolescents assigned to GSC-HIV will demonstrate significantly greater reductions in HIV-risk behaviors than those assigned to TAU. Our second aim is to examine processes of change (i e, mediators) associated with response to the GSC-HIV intervention. Analyses will involve 1) measuring the degree to which participants demonstrate pretreatment/post-treatment changes in selected domains (i e, in attitudes and AOD/risky sex expectancies, relevant social skills, communication skills, decisional balance, self-efficacy) directly reflecting key intervention components, and 2) examining whether changes in these domains predict participants' ultimate response to intervention. A third aim is to examine individual and contextual variables, representing significant subgroups of adolescents that may predict differential treatment response. Potential amenability to treatment factors (i e, moderators) include perceived peer norms for condom use, parent-child relationships and communication, adolescent psychopathology, and psychological maltreatment.